[photopress:face_pic.jpg,full,alignright]If you had to find a computing field which is inherently difficult it is face recognition. Voice recognition was tough enough but there the major barrier was processing power and microphone sophistication. Once those two problems were solved — and in a sense they were self-solving, the basic problems were not that difficult to overcome. Soon voice recognition will be built into mobile phones.
Face recognition is a problem of another order — the difficulties cubed. Now Hewlett-Packard is collaborating with Tsinghua University to develop multimedia technologies including automated photo sorting, facial expression recognition and video and music analysis.
The collaboration brings together two of the world’s leading research institutions in multimedia management. That sounds like a sort of glib marketing statement but this time it is true.
The talent that exists in HP Labs is truly formidable. Its expertise, to be honest, has not been precisely in this area but it is, beyond doubt, the world leader when it comes to handling images. One can see how that expertise could be used to work on face recognition.
Tsinghua University is considered one of China’s leaders in multimedia research, including analysis, process, search and management. Together they may be able to move face recognition to a point where it is seriously useful. Video and music analysis are new world’s to conquer where, as yet, even the problems are not clearly defined. Great and exciting stuff.
Source: DMN Newswire